INSTITUTIONEN FÖR SPRÅK OCH LITTERATURER
AN ANALYSIS OF WAGO/KANGO PREDOMINANCE IN SWEDISH STUDENTS’ JAPANESE VOCABULARY
A cued translation task involving wago and kango synonyms
Marcus Burman
Humanitarian faculty University of Gothenburg
Essay/Thesis: 15 hp Program and/or course: JP1520 Level: Undergraduate Semester/year: Ht/2019 Supervisor: Fusae Takasaki Ivarsson Examiner: Lars Lam
Abstract
Essay/Thesis: 15 hp Program and/or course: JP1520 Level: Undergraduate Semester/year: Ht/2019 Supervisor: Fusae Takasaki Ivarsson Examiner: Lars Larm Keywords: Sino-Japanese words, Kango, Wago, Japanese, Translation task
Aim: The aim of this thesis is to evaluate the lexical type predominance between Japanese-based wago and Chinese-based kango within the vocabulary of Swedish learners of Japanese, comparing those who have studied in Japan and those who have not. Various studies about wago and kango exist, but research regarding bilingual learners with a non-kanji background is still limited.
Theory: The theoretical framework on bilingualism is based on previous research on text entities and their structure as described by De Groot (1992) and Walter (2004). As for the parameters to analyze wago and kango’s basic elements and occurrence, studies conducted by Jin & Yokosawa (2007) and Nakayama (2002) were utilized. The translation task was inspired by Nakayama’s (2002) research on Chinese-Japanese bilinguals.
Method: The translation task was structured to ask participants to fill in suitable translations in the blank for the pronunciation of Japanese words based on the meaning of a corresponding Swedish word. A test consisting of a cued translation task was conducted in order to evaluate lexical type predominance between wago and kango. Intermediate/advanced level Swedish students of Japanese, with or without experience of learning Japanese in Japan, were asked to write two Japanese equivalents (each with a designated number of letters) for each of 40 Swedish words. This without being informed of the true aim of the test, which was to evaluate the wago/kango predominance pattern among the given answers. The two corresponding Japanese translation was acquired from Kanji look and learn (Banno et al., 2009). Preceding the translation task, the participants were instructed to answer a questionnaire to determine their previous experience with Japanese. The participants were not told of the test’s true aim to observe the preference between synonyms of kango and wago but instead were tasked with finding the most suitable translation that would fit inside the highlighted brackets.
Result: The results indicated that the respondents had a predominance to translate words into wago. Hence, within all the 1113 answers, 65% were wago and 35% were kango. Indicating an overall predominance towards wago rather than kango when translating. In addition, no noticeable difference could be observed for participants who had studied in Japan and those who had not.
Acknowledgments
During my years of studying Japanese, both in my home country as well as in Japan, I came across what seemed to be a predominance amongst my peers when speaking and writing for certain types of Japanese words. A large part of mine and my peers’ Japanese vocabulary seemed to consist of words with readings that existed before kanji and adapted after pre-existing native Japanese words, referred to as wago. After a while, however, I noticed that this predominance seemed to disappear the more our knowledge of Japanese increased. This interested me in comparing level-matched second language (L2) learners of Japanese with different study backgrounds and to see what tendencies they exhibit.
First of all, I would like to thank my supervisor Fusae Takasaki Ivarsson, who provided me with guidance and have supported me throughout the duration of this report. Her advice has helped greatly in all stages of my writing, and her quick replies have been greatly appreciated.
I would also like to give my gratitude to my fellow classmate Lisa Bendall, for reading and correcting my paper. Special thanks go as well to Linda for encouraging my research as well as reading and correcting my earlier drafts. I am also grateful to my family and close ones, for encouraging me to pursue my adventures and for their support throughout my years of studying Japanese.
Gothenburg, December 2019
Marcus Burman
3
Table of contents
Acknowledgments ...... 3 1. Introduction ...... 1 2. Theory ...... 1 2.1 wago and kango ...... 1 2.2 Word recognition processing ...... 2 2.3. Previous research ...... 4 3. Presentation of problems ...... 7 3.1. Aim and research questions...... 7 3.1.1. Hypothesis ...... 7 4. Method and Material ...... 8 4.1 Participants ...... 8 4.2 Material ...... 8 4.3 Method ...... 8 5. Result & Analysis ...... 10 5.1 Overall result ...... 10 5.2 Third- semester students’ result ...... 12 5.2.1 Result from data analysis ...... 14 5.3 Fifth- semester students’ result ...... 15 5.3.2 Result from data analysis ...... 16 6. Discussion/Conclusion ...... 17 List of references ...... 20 Appendix I: questionnaire ...... 22 Translation task ...... 22 Appendix II: List of target words and their intended wago and kango equivalents ...... 25 Appendix III: Overall result of the translation task by 20 participants ...... 28
1. Introduction
One of the main questions facing researchers of bilingualism is how different lexical processes are stored and accessed from our long-term memory. De Groot (1992) shows us that a pair of translations may not share meanings completely, but rather the representations of a word’s meaning is distributed over a set of elementary conceptual units. Hence, the two members of a word share a subset of these units at a minimum and each member of a translation pair are associated with several language-specific conceptual units. In the case of a language containing many different lexical entities of foreign origin but with similar meaning, choosing what De Groot (1992) discusses as the most appropriate one, is challenging for most L2 speakers when translating from their native language (L1). Such is the case with Japanese, containing Sino-Japanese words (kango), loan words (gairaigo) and inherently native Japanese words (wago) (Igarashi, 2007). This, of course, poses a challenge for second language (L2) learners of Japanese speakers during translation, when choosing apt lexical entities while still retaining semantical accuracy.
The purpose of this thesis is to study differences in knowledge of the Japanese lexical items wago and kango, focusing on differences between students who have studied in Japan (> than 6 months) or only in Sweden. By observing these differences, hopefully, we are able to study the effects that one’s environment may have on lexical retention.
2. Theory
2.1 wago and kango
Few countries can truly call themselves monolingual, and even supposedly monolingual countries have bilingual and multilingual speakers. This has led to the exchanging of words between languages and Japanese is no exception. The general consensus is that kanji are characters created in China more than 3,500 years ago (Banno et al., 2009). Dating back to the origin of script, the Japanese did in fact not have their own writing system but are believed to have been introduced to kanji via Korea around 500 A.D (Banno et al., 2009). The characters had, however, been made for writing Chinese. Hence, after being introduced to Japan, Japanese kanji evolved to have two different types of pronunciations: kango, Sino-Japanese words that retained its Chinese pronunciation, and wago, readings that existed before kanji and adapted after pre-existing native Japanese words (Igarashi, 2007). Ivarsson (2016) explains the relationship between wago and kango in her introduction to facts and concepts related to kanji (Ivarsson, 2016) by comparing it to the words of Latin/Greek origins and Anglo-Saxon words. ‘The relationship between Sino-Japanese words and native Japanese words in the Japanese vocabulary can be compared to words of Latin/Greek origins and Anglo- Saxon words in the English vocabulary; the Sino-Japanese word sankyaku
1
corresponds to the English word “tripod” (tri“three” + pod “foot” via Latin tripodis from Greek tripous), and the native Japanese mitsuashi to “three-legged” of Anglo-Saxon origin. Sino-Japanese words tend to be used in technical terms and formal expressions, while native Japanese words are often found among basic words and everyday language, analogously to the contrast between words of Latin/Greek origin and those of Anglo-Saxon origin in English.’ (Ivarsson, p. 35, 2016)
A study in 2006 from the National Institute for Japanese Language and Linguistics (NINJAL) illustrates the division of wago and kango in both high school textbooks as well as in children’s reading (Igarashi, 2007). In the study, the percentage of wago in children’s reading (78%) is higher than high school textbooks (40.1%), while at the same time, kango in children’s reading (18.7%) is much lower than in high school textbooks (52.3%) (Igarashi, 2007). Igarashi (2007) reasons that the literature that is intended for children are constructed to have less kanji, thus contributing to a decline in kango, as it is mostly represented through kanji. Conversely, high school literature has a higher occurrence of kango as they have been perceived as more suitable in Japanese academic writing (Igarashi, 2007). Furthermore, this suggests that learners of Japanese should experience similar progress as well, as both Japanese children and second language (L2) learners learn the hiragana script first, and only later are introduced kanji as the last orthographic element to be introduced.
“When students become familiar with hiragana, the number of kanji words increases as the use of hiragana content words decreases.” (Igarashi, p. 40, 2007)
These results are similar to those reported in another report from NINJAL (NINJAL, 2009), as the wago and kango usage in the vocabulary amongst children’s writing was analyzed:
“WAGO, original Japanese words which refer to familiar things and actions, decreased while KANGO, words borrowed from Chinese which primarily represent abstract concepts, increased with grade level.” (NINJAL, p. 182, 2009)
2.2 Word recognition processing
Bilingualism can be defined as regular use of two or more languages (Grosjean, 2015). A person is considered to be bilingual when you are either fluent in both languages (balanced bilinguals), or when you are stronger in one language compared to another (dominant bilingual). It is important to note; however, that although one may be a dominant bilingual, it is rare to have both languages cover all domains of life. In Marian & Spivey’s research (2003), their result indicates that upon hearing two words which sound similar in both Russian and English, the participants’ English and Russian lexical processing system are both activated for bilinguals (Marian & Spivey, 2003). It is understood that the frequency of the word plays a key factor in word choice (Marian & Spivey, 2003). This has been
2
supported in research by Beauvillian & Grainger (1987), arguing that the internal property of the lexicon and the coded frequency of different readings play an important role in lexical access.
In 1992, De Groot defined in her “distributed conceptual feature model”, the determinants of word translation from the subject’s L1 and L2 and where two lexical entities from two languages share different semantic representations. She argues that it is likely that words used often in monolingual settings also occur relatively often in translation settings, which will strengthen the memory connection for representational units between the two translations. As mentioned above, she concluded that a pair of translations may not completely share meanings, but the representation of word meanings is distributed over a set of elementary conceptual units (De Groot, 1992). This re-enforces the theory that a person’s knowledge of these distributions is seen as an important factor in word translation and association during code-switching. Additionally, De Groot’s research indicated that high-frequency words have fewer translation errors than low-frequency words in a monolingual setting.
Walter (2004) goes further on about the transferability of reading comprehension skills for L2, linking it to mental representations of text and L2 working memory. According to Walter:
“[the] working memory (henceforth WM) is a system of mechanism by which humans process the information they need for the performance of complex cognitive tasks and maintain it in an accessible form (…) what is processed and stored can be information from long term memory, or new information, or both.” (Walter, p. 318, 2004)
Walter (2004) and De Groot’s (1992) results indicate that a person’s translation choice is determined by multiple factors, such as frequency and the associations two translation pairs share.
Regarding Kanji recognition and retention, Ivarsson (2016) gives more details into the cognitive model of kanji retrieval. According to Ivarsson (2016), it is a widely acknowledged concept that cognition of a kanji character is supported by the knowledge of other related kanji characters. In circumstances where a character has not yet been mastered, one’s kanji cognition system process the unknown character, and use the partial information known in order to reconstruct sufficient enough information to guess the whole image of the character (Ivarsson, 2016). She further elaborates on the kanji’s storage in one’s mental lexicon, categorizing it into three storages: Form (orthographically similar assembly), sound (phonologically identical assembly, such as ka: 火、下) and meaning (semantically related assembly, words within a similar category of words, “plants”: 木、米、竹). These three categorizes work together for each kanji stored in the learner’s mental lexicon, forming an association network when the number of acquired characters increases (Ivarsson, 2016). For example, in what Ivarsson denotes as the “sound” assembly, words that are phonologically identical (homophonous) will be gathered, such as on-reading ka (日, 火, 下, 何, 夏, 歌, 家, 過, 荷, 化, and 果)
3
and the kun-reading hana (鼻), whereas the synonymous and antonymous characters would be placed in the “semantically related assembly” (Ivarsson, 2016). Regarding a writing task, it is by utilizing the phonological and/or semantic representations of a word that a visual representation can be reconstructed. Hence, an error occurs in this when the reconstruction fails (Ivarsson, 2016). Concerning a person whose L1 language is under the alphabetic writing system (i.e., an orthographically different writing system than Japanese’s kanji), it is acknowledged that the speaker will most likely try to apply their L1 reading strategy when reading kanji (for example, Swedish) (Ivarsson, 2016).
2.3. Previous research
Although there have been some studies that have analyzed Japanese wago (words with Japanese origin) and kango (Sino-Japanese words) knowledge, most of them have either focused on usage amongst L1 Japanese speakers or Chinese-Japanese bilinguals (Nakayama, 2002) (Jin & Yokosawa, 2007). Their research examined words identical in the Chinese and Japanese language, cognates, at both graphemic and semantic level (e.g., Nakayama, 2002). The current framework for the experiment is inspired by the methodology of Nakayama (2002). In her research on “the lexical status effect on cognates”, Nakayama (2002) utilized a word-fragment completion task (WFC task) in order to examine the Chinese-Japanese bilinguals’ lexical processing. She interviewed 22 bilingual Chinese students, all native Mandarin speakers, and late beginning Chinese-Japanese bilinguals. The test items ranged in length from four to six hiragana characters and one of the characters was represented by a space. Each test-item was chosen so that the missing character could be replaced with at least one other character that would still create a word. The participants were first taught 60 study items, and after doing a series of filler problems, tasked with completing fragments of the studied test items. Afterwards, they were asked to rank their recognition of said study items in order to observe the priming effect of cognates on kanji recognition (Nakayama, 2002). Her results indicated that it was not vital that the pronunciation of Japanese and Chinese cognates had to be identical for the participants’ recognition, but rather, it seemed to be vital that the orthographic representations of the cognates were important for the participants’ recognition.
4
Figure 1 Example of the stimuli task in Nakayama's (p. 187, 2002) experiment
Similarly, in 2007, Jin & Yokosawa also studied the attributes of language-switching in Chinese- Japanese bilinguals’ word recognition. Focusing on the comprehension of visually presented words, their research analyzed how the similarities in their writing systems affect cognitive processing by Chinese-Japanese bilinguals. Their result indicated that the tested bilinguals performed quicker in the recognition test when the representation of kanji was identical in both Chinese and Japanese (Jin & Yokosawa, 2007), e.g., words that are cognates (same meaning and spelling in Chinese and Japanese), such as kenkou which has the same kanji representation in both Japanese (kenkou:健康) and Chinese (kenkou:健康) had a higher accuracy amongst the bilinguals. In their experiment, four types of two- Chinese-character (Kanji) compound words were used:
“(1) words that are identical in Chinese and Japanese at both graphemic and semantic levels (cognates); (2) words specific to Japanese which do not have semantic meaning in Chinese (Japanese words); (3) words specific to Chinese, which do no [sic] have meaning in Japanese (Chinese words); (4) non-words in either language (pseudo-words).” (Jin & Yokosawa, p. 1, 2002)
They discussed that lexical processing in both Chinese and Japanese aids each other during word recognition (Jin & Yokosawa, 2007). Hence, words in two languages that had similar orthographic and semantic representations in both languages (for example kango) activated both the Chinese and Japanese lexical processes at the same time, which resulted in faster responses from the bilingual participants (Jin & Yokosawa, 2007). There are several similarities between Nakayama’s (2002) and Marian & Spivey’s (2003) research, which examined spreading activation and lexical processing for bilinguals. Marian & Spivey (2003) used eye-motion detectors to see which objects the participants would focus on when hearing words that sounded phonologically similar but had different meanings in English and Russian. The experiments’ results show that English-Russian bilinguals had a simultaneous activation in their lexical processing systems when English and Russian words were overlapping with their pronunciation (Marian & Spivey, 2003), supporting Jin & Yokosawa’s (2007) and Nakayama’s (2002) results.
5
As mentioned previously, there is little published data on similar research on learners of Japanese with a non-kanji background (i.e. whose L1 writing system is not logographic). Therefore, this research will primarily focus on Swedish L2 speakers of Japanese that will be translating from their first language (denoted as L1).
This research will examine the following two propositions: 1) based on the unequal distribution of wago and kango among the words highlighted to be memorized in the textbook used by all participants, “Kanji look and learn” (Banno et al, 2009), a significant difference may occur between the participant’s knowledge of kango and wago words. 2) The longer the exposure to the Japanese language in Japan, the better the overall knowledge of Japanese words should be for the participants. Students having spent a longer time in Japan will most likely have had more exposure to all types of vocabulary, and therefore, have a more balanced distribution of wago and kango in their vocabulary. Therefore, have a higher accuracy of translations for both kango and wago words.
6
3. Presentation of problems
As explained earlier, previous studies of wago and kango have not dealt with the effects it has on Swedish L2 Japanese speakers. Hence, the present study will aim to see if there is a discrepancy between the distribution of wago and kango usage for non-native Japanese speakers when translating from their native language, which for this particular study would be Swedish.
3.1. Aim and research questions. The aim of this research is to investigate whether Swedish students tend to favor wago over kango when translating (vice versa), and thereafter decipher what sort of tendencies they exhibit.
The research questions are as summarized:
Is there a lexical type predominance of wago and kango within the vocabulary of Swedish learners of Japanese? o If so, what kind of differences in the degree of predominance is observed according to the learners’ study history (length of formal education, the experience of study in Japan) and lexical familiarity?
3.1.1. Hypothesis For the above-mentioned questions, the following hypotheses are constructed:
1. Based on the research of Igarashi (2007), the participants will most likely tend to translate the Swedish words into corresponding wago. Following Igarashi’s (2007) argument which suggests that non-native Japanese speakers should experience a similar language acquisition as Japanese children, learning primarily wago and only later being introduced to kango. This is further supported when looking at the distribution of wago and kango (the former being greater than the latter) in the kanji and grammar/conversation textbooks used by all participants (see Table 5.). 2. The expected result from the comparison between the students who have studied in Japan or only in Sweden is that the former exchange students will have acquired a larger vocabulary of kango. Their answers will, therefore, include a higher ratio of kango than the answers by the students who have only studied in Sweden.
7
3. Method and Material
A test consisting of a cued translation task was conducted in order to evaluate the lexical type of predominance between wago and kango in the vocabulary of Swedish learners of Japanese.
4.1 Participants
20 Swedish students participated in the experiment. All the participants were native Swedish speakers from a large Western university enrolled in either their third- or fifth- semester course of their Japanese bachelor’s degree. At the time of the experiment, 15 of the participants had received formal education in the Japanese language in Japan, whereas 6 of these had had approximately six months of formal education, and 9 had had approximately 12 months. 7 out of 20 participants were enrolled in the third- semester of Japanese language courses, an early intermediate level, while the remaining participants were enrolled in their fifth- semester of Japanese language course, which is at an early advanced level. The participants are all between the ages of 20 and 30, which should allow the difference of age to have less of an impact on the result. All participants had prior to the testing, during their first and second semester, studied the textbooks Kanji Look and Learn (“KLL”) (Banno, et al., 2009), Genki I (Banno, et al., 2011a), Genki II (Banno, et al., 2011b), having learned approximately 450 kanji characters during this period.
4.2 Material
The task was to write two Japanese equivalents for each of 40 Swedish words with hiragana (e.g. もり mori and しんりん shinrin for “skog”; eng. trl. “Forest”). The Swedish words were those that have two synonymous wago and kango equivalents found in the vocabulary list in KLL (Banno, et al., 2009). All the test materials consisted of 80 totals of wago and kango words divided into 40 pairs, retrieved from Kanji look and learn (Banno et al. 2009). The equivalents ranged in length from two to seven hiragana characters. Thirty-five of the equivalents (26 wago and 9 kango) also appeared in the combined vocabulary lists found in Genki I (Banno, et al., 2011a) and Genki II (Banno, et al., 2011b).
4.3 Method
The participants were tested individually online, in the form of a questionnaire and a translation task in a Microsoft Word file. They were told that they would be tested on their vocabulary knowledge, particularly, of Japanese synonyms. For the task, the kanji characters were ignored and instead, empty boxes for writing the Japanese equivalents of the Swedish word using Hiragana characters were shown. Each mora (quasi-syllabic unit in the Japanese sound system) represented one empty box (e.g. 2 boxes
8
for もり and 4 boxes for しんりん). The vocabulary knowledge of the participants was examined in a translation task, where the participants were asked to translate from their native tongue (Swedish) to Japanese. There were two types of cues in this task. The first type was the number of empty boxes indicating the number of hiragana characters to write the intended equivalent, which applied to all of the intended equivalents. The other type was limited to the intended equivalents that are adjectives or action nouns, to which an extra cue of endings was presented (i.e. the last mora -い -i was presented in case of an i-adjective, -な -na for a na-adjective and -する -suru for an action noun). See examples in Table 1. The test list consisted of 80 Japanese items and 40 Swedish items. In Nakayama’s research (2003), they perform recognition tasks and WFC tasks to see the priming effect of cognates on Japanese-Chinese bilinguals. In this paper, however, the exercises have been constructed in such a way that the participants will not be allowed to study the terms before the tests, as to not prime them on their word choice.
Table 1 Examples taken directly from the translation task (see Appendix I for complete list). The intended words taken from KLL (Banno et. Al, 2009) are (upper row) あたたかい (wago), and おんだん な (kango), as well as (lower row) かわる (wago) and へんかする (kango).
Tecken 1 Tecken 2 Tecken 3 Tecken 4 Tecken 5 Tecken 6 Tecken 7 Varmt 1. い 2. な
In the distributed survey, a Swedish word was presented with two slots for typing in two equivalents of the designated number of hiragana in a word file (see Appendix I for complete list). The participants were asked to write in hiragana within the boxes highlighted, which indicated the number of syllables the translation should have (but not restricted to). The participants were also asked to complete the task within 10 to 15 minutes. The participants could return to previous questions if need be.
Table 2 Illustration of the translation task. The test items contain at the end a “translation cued- fragment” (see APPENDIX I for complete list)
LEXICAL ITEM JAPANESE ITEM TEST ITEMS (SWEDISH) - NOT SHOWN - - SHOWN - WAGO (VARMT) あたたかい（暖かい） い
KANGO (VARMT) おんだんな（温暖な） な KANGO (“NÅGOT FÖRÄNDRAS”) へんかする（変化する） す る
9
5. Result & Analysis
In the following sections, results from all participants in the translation task will be presented in subsection 5.1, while further analysis of the differences among the third- and fifth-semester students will be made in subsections 5.2 and 5.3 respectively. For a complete list of all translations and the number of translations, see Appendix III. For both the third- and fifth-semester students, a two way between subjects ANOVA was conducted to compare the effects of time spent studying in Japan to one’s choice of wago and kango.
In the result of the translation task (see Appendix III), the answers were separated into three categories. The first category is the intended equivalents (“Varmt”’s intended equivalent would be あ たたかい), the second group are plausible equivalents with a synonymous meaning to the intended equivalent (warm; あたたかい and warm; あつい are synonyms) and answers that have an incorrect spelling and/or cannot semantically be used in the same way as the intended equivalents (for example あただかい would be incorrect).
Table 3 Example of different types of answers
SWEDISH WORD INTENDED SYNONYMOUS INCORRECT EQUIVALENT EQUIVALENT VARMT あたたかい あつい あただかい
The focus of this research is not to decipher if the participants’ answers will match the intended equivalents, but the overall usage of wago and kango. In this study, we will only be looking at the intended and the plausible equivalents.
5.1 Overall result
In Figure 2, the sum of all wago and kango answered by the 20 participants will be presented.
10
LEXICAL TYPE DIFFERENCES
Matching highlighted squares Correct, not matching highlighted squares
71
656 52 334
WAGO KANGO
Figure 2 The sums of wago and kango equivalents from all participants
Out of the participants, 15 had received formal education in the Japanese language in Japan, whereas 6 of these had had approximately sixth months of formal education, and 9 had had approximately 12 months (also including overall time spent in the country or on vacation). Due to the participants mainly leaving for their exchange studies during the beginning of their third- semester, none of the participants in their third- semester of Japanese had studied and/or been in Japan for more than 6 months at the time of the experiment. The material for the experiment was taken directly from KLL (Banno, et al. 2009). Overall there were 71 wago and 52 kango for a total of 123 answers amongst the participants that were synonyms for intended translations, i.e, could be interchanged and retain the same meaning in a sentence (for example, Frukost (Breakfast): あさごはん (intended); あさめし). If all participants had perfect answers (as intended), there would have been 20 participants multiplied by 40 words which equal 800 wago and 800 kango, but there were 727 wago (89% of intended) and 386 ango (ca 48% of intended). Hence, within all the 1113 nswers, 65% were wago and 35% were kango. Indicating an overall predominance towards wago rather than kango when translating.
When analyzing the ratio of wago to kango occurrences, each category of groups had a clear overrepresentation of wago (shown in Table 4).
11
Table 4 Ratio of wago to kango words as it appears in each group
Only in 0-6 months 6-12 months Participants Sweden in Japan in Japan
5th semester 185% 188% 184%
3rd semester 248% 212% N/A
As can be seen from Table 4, all of the items had a value above 100%, which indicates that all groups’ answers had a higher ratio of wago than kango (the value is gained from dividing the sum of intended and plausible wago with the sum of intended and plausible kango). The third-semester students who had not had any studies abroad in Japan produced the largest difference of wago to kango, 148% more wago than kango, whereas the fifth-semester students who had been studying in Japan between 6-12 months produced the lowest difference, 84%.
It can also be observed that the longer the period time that the participants had studied Japanese, the smaller the difference between the ratio of wago and kango translations become amongst the third- semester students, decreasing by 16 percent from 248% for the third-semester students in Sweden to 212% for the students that had studied in Japan for 6 months. This trend, however, was not exhibited by the fifth-semester students, retaining an even ratio amongst all three groups around 186%.
5.2 Third- semester students’ result
Out of the third-semester students, four had received formal education in the Japanese language in Japan, whereas the remaining three had had approximately sixth months of formal education (also including overall time spent in the country or on vacation). As previously stated, due to the participants mainly leaving for their exchange studies during the beginning of their third- semester, none of the participants in their third-semester of Japanese had studied and/or been in Japan for more than 6 months at the time of the experiment. The results of the translation task for the third- semester participants are shown in Figure 3. In the experiment, the students in their third- semester translated 237 times from the Swedish words into wago, and 105 times into kango. If all seven third- semester participants had perfect answers (as intended), it would be seven participants times 40 words, equivalent to 280 wago and 280 kango, but there were 237 wago (85% of intended) and 105 kango (ca 37% of intended). Hence, within the 342 answers, 69% were wago and 31% were kango.
12
LEXICAL TYPE DIFFERENCE THIRD SEMESTER STUDENTS
Matching highlighted squares Correct, not matching highlighted squares
16
221 12
93
WAGO KANGO
Figure 3 The sums for wago and kango equivalents from third- semester students
In Figure 3, the average of the participants’ answers is shown as a function of the time spent abroad in Japan studying.
In terms of variation of answers, the answers which all of the students answered was: あさごはん (hirugohan, “frukost”)、あたたかい (atatakai, “varmt”)、あぶない (abunai, “farligt”)、おとこ (otoko, “man”)、おや (oya, “föräldrar”), おわり (owari, “slut”)、おんな (onna, “kvinna”)、かう
(kau, “att köpa”)、かわ (kawa, “flod”)、くるま (kuruma¸”bil”)、けす (kesu, “att radera”)、たつ (tatsu, “att ställa sig upp”)、たべもの (tabemono, “mat”)、つかう (tsukau, “att använda”)、ともだ ち (tomodachi, “vän; vänner”)、ゆうじん (yuujin, “vän; vänner”), ひるごはん (hirugohan,
“lunch”)、もり(mori, “skog”) and わかもの (wakamono, “barn”). Out of these 19 words, 18 are wago and 1 is kango. Additionally, the four words no participant answered on was しょうきょする (shoukyo-suru, “att radera”), きりつする (kiritsu-suru, “att ställa sig upp”), じゃくはい (jakuhai, “unga personer”) and きゅうそくする (kyuusokusuru, ”att vila”). All four of these are kango.
If we now turn to the differences amongst the students who studied in Japan and students who only studied in Sweden, based on the average answers from the groups, the students that had studied in Japan had a higher score rating than the students who lived only in Sweden in kango (see Figure 4).
13
5.2.1 Result from data analysis
AVERAGE ANSWERS FOR THIRD SEMESTER STUDENTS
60,00 50,00 40,00 30,00 20,00
10,00 CORRECT CORRECT ANSWERS 0,00 0 m 6 m kango 13,33 16,25 wago 33,00 34,50
Figure 4 The distribution of wago and kango words between third- semester students
The mean score for the three students with no formal education in Japanese abroad was 33 wago and approximately 35 for the four students with 6 months of studies abroad (a 4,3% difference). For the kango, the average score was 13 for students with no formal education in Japan and 16 for students who had studied 6 months in Japan (a difference by approximately 23%). Based on the average response for each category, there was only a 22% difference between the highest and lowest response of kango between the two groups and a 4,5% difference between the highest and lowest response of wago (depicted in Figure 6). Individually, the participant who had the highest number of correct equivalents had out of 80 answers 56 correct (34 wago and 22 kango), whereas the participant with the lowest number of correct translations (both intended and plausible equivalents) had 39 translation (31 wago and 8 kango). Both aforementioned participants belonged to the category of students who had had no experience studying abroad. Furthermore, a two way between subjects’ ANOVA was conducted to compare the effects of one’s duration (0 months and 6 months) spent in Japan to the number of translations of wago and kango. Due to the unequal variance of sample size, six of eight random samples from the six months participants were used in order to minimize the type 1 errors.
Data analysis of variance (ANOVA, with a p=.05) revealed that there was no significant effect of duration spent in Japan on choice of lexical type wago or kango for the third-semester students. F(1, 8) = 1.389, p=0.272.
14
5.3 Fifth- semester students’ result Turning now to the fifth- semester students, a similar trend as shown in Figures 3 and 4 can be observed in Figures 5 and 6. In the experiment, the students in their fifth- semester translated 490 times from the Swedish words into wago, and 265 times into kango. If all thirteen fifth-semester participants had perfect answers (as intended), there would be thirteen participants times 40 words, equivalent to 520 wago and 520 kango, but there were 490 wago (ca 94% of intended) and 265 kango (ca 50% of intended). Hence, within the 755 answers, 65% were wago and 35% were kango.
LEXICAL TYPE DIFFERENCE FIFTH SEMESTER STUDENTS
Matching highlighted squares Correct, not matching highlighted squares
55
41 435
224
WAGO KANGO
Figure 5 The sums of wago and kango equivalents from the fifth- semester students
In terms of variations of answers, the words which each of the fifth- semester students answered was: あさごはん (hirugohan, “frukost”)、あたたかい (atatakai, “varmt”)、あぶない (abunai, “farligt”)、いたみ (itami, “smärta”)、おわり (owari, “slut”)、おんな (onna, “kvinna”)、かう (kau,
“att köpa”)、かわ (kawa, “flod”)、くだもの (kudamono, “frukt”)、くるま (kuruma¸”bil”)、けす
(kesu, “att radera”)、しあわせな (shiawasena, “lycklig”)、たつ (tatsu, “att ställa sig upp”)、たべもの (tabemono, “mat”)、つかう (tsukau, “att använda”)、しつもん (shitsumon, “fråga”)、ともだち (tomodachi, “vän; vänner”)、ひるごはん (hirugohan, “lunch”)、むし (mushi, “insekt”) and わかも
の (wakamono, “barn”). Out of these 20 words, 19 are wago and 1 is kango. Additionally, like for the third- semester students, the four words no participant answered on was しょうきょする (shoukyo- suru, “att radera”), きりつする (kiritsu-suru, “att ställa sig upp”), じゃくはい (jakuhai, “unga personer”) and きゅうそくする (kyuusokusuru, ”att vila”). All four of these are kango.
15
5.3.2 Result from data analysis
AVERAGE ANSWERS FOR FIFTH SEMESTER STUDENTS 70,00 60,00 50,00 40,00 30,00 20,00
CORRECT CORRECT ANSWERS 10,00 0,00 0 m 6 m 12 m kango 20,00 20,00 20,56 wago 37,00 37,50 37,89
Figure 6 The average of the answers for the fifth- semester students as a function of time spent abroad in Japan (in months)
The response amongst the fifth- semester students was similar to the third- semester students as the result indicating a predominance towards translating to wago. Based on the average response for each category, regardless of their study background, there was only a 3% difference between the highest and lowest response of kango and the highest and lowest response of wago (depicted in Figure 6 and Table 4). Individually, the participant who had the highest number of correct equivalents had out of 80 answers 72 correct (39 wago and 33 kango), whereas the participant with the lowest number of correct translations (both intended and plausible equivalents) had 38 translation (36 wago and 2 kango). Both participants belonged to the category of students who had studied abroad for 6-12 months. Furthermore, like previously, a two way between subjects ANOVA was conducted to compare the effects of one’s duration (0 months, 6 months and 12 months) spent in Japan to the number of translations of wago and kango. Due to the unequal variance of sample size, only four samples were taken from the group of students with 12 months of time spent in Japan.
Data analysis of variance (ANOVA, with a p=.05) revealed that there was no significant difference between duration spent in Japan and choice of lexical type wago or kango for F(2,6) = 1.131, p=0.383.
16
6. Discussion/Conclusion
The current study aimed to determine if there is a predominance of wago over kango in the vocabulary of Swedish learners of Japanese. In addition, to observe what tendencies are exhibited between participants who have studied Japanese in Japan or only in Sweden.
As could be seen from the overall result, participants from each subgroup tended to favor wago translations over kango translations (see Figures 2, 3 and 5). Without making a distinction if the participants had at one point studied in Japan or not, if all participants had perfect answers (as intended), there would have been 20 participants multiplied by 40 words which equal 800 wago and 800 kango, but there were 727 wago (89% of intended) and 386 kango (ca 48% of intended). Hence, within all the 1113 answers, 65% were wago and 35% were kango. These findings, therefore, indicate an overall predominance towards wago rather than kango when translating.
When comparing the participants who have studied in Japan and those who have only studied in Sweden, it was, however, revealed that only a minor difference could be observed. Surprisingly, regardless of their time spent in Japan, the ratio of wago to kango answers for the fifth- semester students were seemingly even, with an approximately 186% ratio of wago to kango for each group (see Table 4). For the third- semester students, as Table 4 shows, there is a difference: 212% ratio of wago and kango usage for participants that have studied in Japan and 248% for those who have not. It is important to note that when looking at the fifth- semester students, regardless of their time spent in Japan, a noticeable difference between their usage of wago and kango could not be observed, contradictory to the third- semester students. This is further supported by the ANOVA analysis, indicating no significant difference between the groups for the duration of time spent in Japan. Therefore, the findings in this study have been unable to prove the hypothesis as it could not demonstrate that studying in Japan can contribute to a difference in the students’ usage of wago and kango.
Interestingly, among the words that are prioritized to be learned first in KLL (Banno, et al., 2009), there was an overwhelming amount of wago words being utilized.
Table 5 Distribution amongst the chosen wago-kango pairs in KLL (Banno et al. 2009) and Genki I (Banno, et al., 2011a) and Genki II (Banno, et al., 2011b).
WAGO KANGO WORDS TO BE MEMORIZED IN KLL 78% 25% (BANNO ET AL., 2009) APPEARANCE IN GENKI I AND II 70% 25% (BANNO, ET AL., 2011A) (BANNO, ET AL., 2011B).
17
Out of 40 wago words used, 78% of the intended equivalents (see Table 5) were marked as priority words, denoted with grey marking within the book. The corresponding percentage for kango words was 25%. Conversely, the most frequent equivalents were all from the study material which appears in both KLL (Banno, et al, 2009) and the combined glossary of Genki I (Banno, et al., 2011a) and Genki II (Banno, et al., 2011b). There were 14 out of the 80 intended equivalents which each of the 20 participants translated to. Each of these equivalents appears in the glossary in Genki I (Banno, et al., 2011a) and Genki II (Banno, et al., 2011b), (see Appendix II and III.). The equivalents which 19 of the participants chose were also words that appeared in the glossary in Genki I (Banno, et al., 2011a) and Genki II (Banno, et al., 2011b). These findings are consistent with the reasoning of Igarashi (2007). As mentioned earlier, Japanese language learners will typically have a higher usage of wago, due to their early exposure of words written in hiragana (Igarashi, 2007). However, as their studies progress, their increasing usage of kanji should resolve in an increase of exposure to kango (as seen in the comparison between the third- and fifth- semester students, as the longer the progression of studies in Japanese, the less difference occurs between their translation into wago and kango).
This paper set out to investigate two questions: is there is a lexical type predominance of wago and kango within the vocabulary of Swedish learners of Japanese? And if so, what kind of differences in the degree of predominance is observed according to the learners’ study history (length of formal education, experience of study in Japan) and lexical familiarity? In this paper, the test proved the first hypothesis that participants will most likely tend to translate the Swedish words into the corresponding wago as it could identify a tendency of predominance for wago translations for Japanese L2 speakers when translating from Swedish. The findings in this study have, however, been unable to give any conclusive evidence for the second hypothesis as it could not demonstrate that studying in Japan can contribute to a difference in the students’ usage of wago and kango.
However, it is important to note, that this study may contain bias as the research had primarily focused on the translation capabilities and word associations between the participants’ Swedish lexicon and Japanese lexicon, instead of the depth of word knowledge.
At the beginning of this experiment, there was a time interval since some of the participants had last been to Japan. However, this factor did most likely not play a major role in affecting the result as the now mentioned participants were either currently in Japan and a minor period of time had passed from their return to the day they performed the task, all whilst having actively studied Japanese at their home university.
The relatively limited sample notwithstanding, this work offers valuable insight into the translation preferences for wago and kango for non-native speakers of Japanese with a non-kanji background.
18
For future reference, research questions that could be asked include how similar participants would act if they were asked to do a sentence completion task instead of a cued translation task. A task that could be asked of the participants could be to fill in the appropriate wago and kango equivalents, based on the context of the sentence, (e.g. according to Igarashi (2007), “breakfast”, a wago equivalent 朝ごは ん asagohan is produced in a more informal/colloquial context and a kango equivalent 朝食 choushoku for a formal/written language context). This would allow researchers to determine the depth of knowledge of the lexical entities and hopefully provide a different account on the general knowledge of wago and kango words used in a practical scenario.
19
List of references
Banno, E., Ikeda, Y., Ohno, Y., Shinagawa, C. and Tokashiki, K. (2011a). Genki: an integrated course in elementary Japanese (second edition). Tokyo: The Japan Times.
Banno, E., Ikeda, Y., Ohno, Y., Shinagawa, C. and Tokashiki, K. (2011b). Genki II: an integrated course in elementary Japanese (second edition). Tokyo: The Japan Times.
Banno, E., Ikeda, Y., Shinagawa, C., Tajima, K. and Tokashiki, K. (2009). Kanji look and learn, 512 kanji with illustrations and mnemonic hints Tokyo: The Japan Times.
Beauvillian. C and Grainger. J, (1987). Accessing interlexical homographs: some limitations of a language-selective access. URL: https://search.proquest.com/docview/1297342167?pq- origsite=gscholar
De Groot. A. M. B. (1992). Determinants of Word translation, university of Amsterdam, Amsterdam, The Netherlands. Journal of Experimental Pshycology; Learning, Memory; and Cognition 1992. Vol. 18. No. 5. 1001-1018. URL: https://www.researchgate.net/profile/Annette_Groot/publication/232521607_Determinants_of_Word_Tran slation/links/56eac2c308aec6b500162e24/Determinants-of-Word-Translation.pdf
Grosjean, F. (2014). Bicultural Bilinguals. International journal of bilingualism 2015, vol. 19. URL:https://journals.sagepub.com/doi/pdf/10.1177/1367006914526297?casa_token=Jblamg7ghsQAA AAA%3AEbrruugywga2HR9QahiSSHvtPLRUeqNhwFXEEdfVM7_5Ld6yNxF2MZU49Cmlk632f2i p17hERxPRFw&
Igarashi, Y. (2007). The Changing Role of Katakana in the Japanese Writing System: Processing and Pedagogical Dimensions for Native Speakers and Foreign Learners, University of Victoria (Canada), ProQuest Dissertations Publishing, 2007. NR41190. URL: https://dspace.library.uvic.ca:8443/bitstream/handle/1828/189/PhD%20dissertation.pdf?sequence=1&i sAllowed=y
Ivarsson, F. (2016). “A study of L2 kanji learning process, Analysis of Reading and Writing Errors of Swedish Learners in Comparison with Level-matched Japanese Schoolchildren.” University of Gothenburg, Print: Repocentralen, Campusservice Lorensberg.
Jin, H. and Yokosawa, K. (2007). The attribute of language-switching in Chinese-Japanese bilinguals’ word recognition, The Japanese journal of psychonomic science 2007, vol. 25, No. 2, 279-280)
Marian, V. & Spivey, M. (2003). Competing activation in bilingual language processing: Within- and between-language competition. Language and cognition 6 (2). URL: https://pdfs.semanticscholar.org/e993/917a4eb1c72099436822d5f61f4655bef9d2.pdf
Nakayama, M. S. (2002) The cognate status effect in lexical processing by Chinese-Japanese bilinguals, psychologia, 2002, 45, 184-192.
National Institute for Japanese Language and Linguistics, 2009. “Vocabulary used in children’s writing” Report 98, 1989. Tokyo Shoseki Co., Ltd. An introduction to the National institute for
20
Japanese language: A sketch of its achievements fifth- edition (2009). URL: http://doi.org/10.15084/00001585.
Walter, C., (2004). Transfer of reading comprehension skills to L2 is linked to mental representations of text and to L2 working memory. Oxford university press, applied linguistics 25/3: 315-339.
21
Appendix I: questionnaire
Translation task Complete list translation task the participants performed.
Example Tecken Tecken Tecken Tecken Tecken Tecken 1 Tecken2 3 4 5 6 7 Varje år 1. ま い と し
2. ま い ね ん
Annat att tänka på っ litet tsu räknas som ett tecken ゃ、 ゅ、 Små ゃ、ゅ、よ räknas inte som egna tecken (ちゅうごく= 4 よ tecken)
Svenska Japanska Tecken Tecken Tecken Tecken Tecken Tecken Tecken 1 2 3 4 5 6 7 Frukost 1.
2. Varmt 1. い 2. な Farligt 1. い 2. な Krig 1. 2. Smärta 1. 2. Att skicka (ett meddelande) 1. 2. す る Man 1. 2. Föräldrar 1. 2. Slut 1.
22
2. Kvinna 1. 2. Att köpa 1.
2. す る Flod 1. 2.
2. Att radera 1. 2. す る Svar 1. 2. Lycklig 1. な 2. な Att ställa sig upp 1.
2. す る Att se fram emot; att hoppas 1. に す る på 2. す る Resa 1. 2. Mat 1.
2. Skillnad 1.
23
2. Att använda 1. 2. す る En fråga 1. 2. Vänner 1. 2. Namn 1. 2. Lunch 1. 2. Inte alls 1. 2. Att bjuda in
2. Skog 1. 2. Insekt 1. 2. Att vila 1. 2. す る Att korsa 1. 2. す る Unga personer 1. 2. Allt (saker) 1. 2.
24
Appendix II: List of target words and their intended wago and kango equivalents
All wago and kango equivalents have been taken from KLL (Banno, et al, 2009). Those in bold letters are marked to be memorized in the book. The words here are presented in katakana, Kanji and then romaji.
Swedish Intended wago Intended kango Appear in the glossary equivalents equivalents of Genki I/Genki II (Banno, et al, 2011a, 2011b) 1. Frukost アサゴハン チョウショク アサゴハン 朝ごはん 朝食 asagohan choushoku 2. Varmt アタタカイ オンダンな アタタカイ 暖かい 温暖な atatakai ondan-na 3. Farligt アブナイ キケンな アブナイ 危ない 危険な Abunai kiken-na 4. Krig イクサ センソウ センソウ 戦 戦争 Ikusa sensou 5. smärta イタミ クツウ 痛み 苦痛 Itami kutsuu 6. att skicka (ett オクル ソウシンする オクル meddelande) 送る 送信する Okuru soushin-suru 7. Man オトコ ダンセイ オトコ 男 男性 Otoko dansei 8. Föräldrar オヤ リョウシン オヤ 親 両親 リョウシン Oya ryoushin 9. slut オワリ シュウリョウ オワル 終わり 終了 owari shuuryou
10. Kvinna オンナ ジョセイ オンナ 女 女性 Onna josei 11. Att köpa カウ コウニュウする カウ 買う 購入する Kau Kounyuu-suru
25
12. flod カワ カセン カワ 川 河川 Kawa kasen 13. något förändras カワル ヘンカする カワル 変わる 変化する Kawaru henka-suru 14. Att tänka på カンガエル シコウする カンガエル 考える 思考する kangaeru shikou-suru 15. Att bestämma キメル ケッテイする キメル (något) 決める 決定する Kimeru kettei-suru 16. Apotek クスリヤ ヤッキョク 薬屋 薬局 kusuriya yakkyoku 17. Frukt クダモノ* カジツ 果物 果実 kudamono Kajitsu 18. Bil クルマ ジドウシャ クルマ 車 自動車 Kuruma jidousha 19. Att radera ケス ショウキョする 消す 消去する Kesu shoukyo-suru 20. Svar コタエ カイトウ コタエ 答え 回答 Kotae kaitou 21. Lycklig シアワセな コウフクな シアワセな 幸せな 幸福な shiawase-na koufuku-na 22. Allt スベテ ゼンブ ゼンブ 全て 全部 Subete zenbu 23. Att Ställa sig upp タツ キリツする タツ 立つ 起立する Tatsu kiritsu-suru 24. Att se fram emot; タノシミにする キタイする タノシミ att hoppas på 楽しみにする 期待する tanoshimi-ni-suru kitai-suru 25. Resa タビ リョコウ リョコウ 旅 旅行 Tabi ryokou 26. Mat タベモノ ショクリョウヒン タベモノ 食べ物 食料品 tabemono shokuryouhin
26
27. Skillnad チガイ ソウイ チガイ 違い 相違 Chigai soui 28. Att använda ツカウ シヨウする ツカウ 使う 使用する Tsukau shiyou-suru 29. Fråga トイ シツモン シツモン 問い 質問 Toi shitsumon 30. Vän; Vänner トモダチ ユウジン トモダチ 友達 友人 ユウジン tomodachi yuujin 31. Namn ナマエ シメイ ナマエ 名前 氏名 Namae shimei 32. Lunch ヒルゴハン チュウショク ヒルゴハン 昼ごはん 昼食 hirugohan chuushoku 33. Inte alls マッタク ゼンゼン ゼンゼン 全く mattaku zenzen 34. Att bjuda in マネク ショウタイする ショウタイする 招く 招待する Maneku shoutai-suru 35. Alla ミナ ゼンイン ミナ 皆 全員 ゼンイン Mina zenin 36. Skog モリ シンリン 森 森林 Mori shinrin 37. Insekt ムシ コンチュウ ムシ 虫 昆虫 Mushi konchuu 38. Att vila ヤスム キュウソクする ヤスム 休む 休息する Yasumu kyuusoku-suru 39. att korsa ヨコギル オウダンする 横切る 横断する yokogiru oudan-suru 40. unga personer ワカモノ ジャクハイ 若者 若輩 wakamono jakuhai *_ are for kanji with an irregular reading
27
Appendix III: Overall result of the translation task by 20 participants
Table 6 - Result of the translation task and variation of translation between the 20 participants
Swedish Answers coherent with the Accurate translations Incorrect word appendix differing from appendix translations
Quantity Quantity Quantity
しょうしょ Wago あさごはん 20 あさめし 4 く 1 Frukost Kango ちょうしょく 8 あたたかい 20 あたたかな 1
Wago あつい 3 Varmt Kango おんだんな 6 Wago あぶない 20 けけんな 1 Farligt kango きけんな 12 いくさ 6 あらそい 1
Wago たたかい 1 Krig Kango せんそう 20 こうせん 1 せんそ 2 Wago いたみ 17 いたい 3 くるし 1 Smärta Kango くつう 6 くやさ 1 とどける/とど Wago おくる 17 く 3 おこる 1 そうしんする 1 へんしんする 2 Att skicka はっそうする 1 (ett meddelande) Kango れんらくする 3 Wago おとこ 19 かれ 1 Man Kango だんせい 15 だんし 1 だんじょう 1 Wago おや 19 Föräldrar Kango りょうしん 18 ふぼ 1 しんせき 1
おわり 17 あいすむ 1 Wago さいご 1 しゅうりょう 7 おしまい 1 さいごう 1 けつまつ 1 おせまじ 1 Slut Kango しゅうまつ 1 Kvinna Wago おんな 20
28
Kango じょせい 16 ふじん 1 かのじょ 1 Att köpa Wago かう 19 かいものする 9 はる 1 Kango こうにゅうする 4 Flod Wago かわ 19 にわ 1 Kango かせん 2
Kango しこうする 2 Att bestämma
29
Att ställa sig upp Wago たつ 20 あがる 1
Kango きりつする 0 たちあがる 1 Att se fram emot; att hoppas på wago たのしみにする 10 ねがう 1 Kango きたいする 3 きぼうする 1 ねがいする 2 Resa Wago たび 16 りょこ 1 Kango りょこう 18 りょうこ 1 Mat Wago たべもの 20 めし しょくりょうひ Kango ん 2 しょくじ 1 しょくひん 1 しょくりょう 2 Skillnad Wago ちがい 14 ちがう 2 ちがさ 1 Kango そうい 4 くべつ 1 くらべ 1 さ 1 さが 1 てんさ 1 Att använda Wago つかう 20 Kango りようする 5 しようする 8
とい 5 きき 1 Fråga Wago だい 1
Kango しつもん 19 もんだい 1
ぎもん 1 Vän; vänner Wago ともだち 20 なかよい 1 Kango ゆうじん 16 しんゆう 1 ゆじん 1 Namn なまえ 19 めいし 1 Wago めいしょう 1 Kango しめい 10 みょうじ 2 みょうし 1 じゅうしょ Lunch Wago ひるごはん 20 ひるめし 5 く 1 しゅうしょ Kango ちゅうしょく 8 く 1 しょうしょ く 1 Inte alls まったく 1 けっして 2 なんにも 1
あまり 1 あんまり 1 なんとも 1 ちっとも 1 Wago ちょっとも 1 Kango ぜんぜん 18 とうてい 1
30
Att bjuda in まねく 2 さそう 15 さそる 1 ゆうはつす Wago よぶ 1 る 1 Kango しょうたいする 7 おごる 1 Alla Wago みなさん 5 みんな 17 ぜんぶ 1 Kango ぜんいん 11 まんにん 1 ぜいん Skog Wago もり 20 はやし 1 りんりん 1 Kango しんりん 10 Insekt Wago むし 19 むしけら 1 こんじゅう 1 Kanog こんちゅう 3 Att vila やすむ/やすみ 15 やすめる 1 りらくする 1 Wago ねむる 2 きゅうけいす Kango きゅうそくする 0 る 6 Att korsa よこぎる 2 こえる 1 よこびる 1 わたる 1 wago まじわる 3 kango おうだんする 3 Unga personer wago わかもの 20 としわか 1 じゅねん 1 kango じゃくはい 0 しょうねん 1 してい 1 せいねん 2
* Words in bold type are a combination of Japanese-based reading (kun’yomi) and Chinese-based reading (on’yomi).
31